The goal of this project is to investigate gait variability in order to generate new insights into the mechanisms of falls and to provide a practical means for improving assessment of fall risk. Using a portable gait monitoring system that the Principal Investigator recently developed, the step-to-step changes in the time spent in consecutive gait cycles, and their sub-phases can, for the first time, be readily measured in clinical settings. Preliminary results demonstrate the potential clinical importance of these measures: in contrast to standard measures of walking (e.g., velocity), step-to-step variability is significantly increased in elderly fallers, as well as in certain subjects with neurological impairment. While a number of factors are known to affect the gait of these subjects, the mechanisms(s) of the increased step-to-step changes in walking (gait instability) and its relationship to fall risk has not been studied. Because gait variability may serve as a sensitive and clinically relevant index of walking ability in the evaluation of intervention trials and fall risk assessment, this investigative team believe it is critical to understand its origins. The long-term goals of this project, therefore, are 1) to quantitatively characterize normal, pathologic and age-related physiologic changes that alter normal step-to-step variability; 2) to identify the mechanism(s) that contribute to normal and abnormal gait variability; 3) to identify interventions for reducing abnormal gait variability and fall risk; and, 4) to develop practical measures of gait instability that can be used to augment clinical gait assessment and to help identify persons at risk of falls. To begin to achieve these long-term objectives, the investigators have identified the following specific aims: A) Test the hypothesis that decreased muscle function, motor control, and aerobic capacity contribute to gait instability (increased gait variability). B) Test the hypothesis that exercise improves gait stability, and C) Test the hypothesis that increased gait variability predisposes the elderly to falls.